We propose to solve two long-standing problems in the field of pre-mRNA splicing: (1) how does the branch site recognition region (BSRR) of U2, which is a single unique sequence, recognize (via base-pairing) a variety of branch site sequences (BSS) in pre-mRNAs during splicing? And (2) how does the pseudouridine (?), which is abundantly present in the U2 BSRR, contribute to this recognition? To address these questions, we recently developed a screening system, allowing us to screen a library of pre-mRNAs with randomized BSS sequences, under various genetic backgrounds where the numbers and combinations of ?s in the U2 BSRR are manipulated and controlled. Our initial screen has generated exciting preliminary results, placing us in a unique position to solve the aforementioned long-standing problems. Three specific aims are proposed. 1. To expand the list of BSS and its 5' adjacent sequences recognized by different U2 variants We will build on our preliminary results and continue to screen the library (randomized BSS). Given that we have also shown in our preliminary experiments that a 6-nucleotide sequence 5'-adjacent to the BSS in pre- mRNA is also recognized by the U2 BSRR, we will screen additional libraries of pre-mRNAs with this 5'- adjacent sequence being randomized, under different U2 (differ in ?) backgrounds. In doing so, we expect to obtain a long list of BSS and its 5' adjacent sequences selected by different U2 variants with different numbers and combinations of ?s in the BSRR, thus helping decode how ?s contribute to BSS recognition. 2. To evaluate whether/how ?s within the U2 BSRR contribute to BSS selection / gene expression With the BSS and its 5' adjacent sequences handy, we will search for the selected sequences in naturally occurring pre-mRNAs. The splicing efficiency of these pre-mRNAs will be tested under U2 variants (differ in ?). In doing so, we can link ? in the U2 BSRR to splicing/gene regulation. We also expect that some U2 variants prefer one BSS+5' adjacent sequence over another, whereas other U2 variants have completely opposite preference between the same two BSS+5' adjacent sequences. We will create several constructs with two BSS+5' adjacent sequences being placed in parallel, and test the usage of one BSS over the other during splicing under specific U2 backgrounds, thus linking ?-mediated BSS selection to alternative splicing. 3. To dissect how individual U2 RNA variants recognize various BSSs and their 5'-adjacent sequences Due to incomplete pseudouridylation at any given site, there exists a mixture of U2 variants (differ in ?) in cells. We will dissect, in detail, how the mixture of U2 variants individually recognizes a specific BSS during splicing. We will use the Oxford nanopore technology to quantitatively map, at single molecule level, ? in the BSRR of U2 isolated from the spliceosomes, which are assembled onto a pre-mRNA with a specific BSS+5' adjacent sequence. In doing so, we can assess how each individual U2 variant behaves, in a natural context, in the process of BSS recognition. A complete picture of ?-mediated BSS recognition is expected to emerge.